REQUIREMENTS

You need perl to run the database software. It is recommended to use at
least perl4, patchlevel 36. Perl5 is untested and there are still several
(unknown) problems with perl5 reported. Bug reports are welcome!

We have tested this software on:

- SUN 4/25 running SunOS 4.1.3 and various other SPARC architectures
- BSDI 
- Linux

It will not run out of the box on systems that don't support 'flock'
function calls like:

- Solaris

However, it is possible to use special hacked versions of perl with
emalated 'flock' support on this systems. If you really want to do this
go ahead but we will not recommend and support this ;-).

You can use any dbm package you want. Note that the software expects to
see index files that have postfixes : .pag/.dir/.db. You might need to
make a (small) change in the software when your dbm package uses other
postfixes then these. Also note that most dbm packages have some
limitations in key and value sizes. The sample configuration file setup
is very conservative but you might need to change some indexing related
values if you experience problems.


INSTALLATION

The best thing to do is make a special tree for this database package.
Make a directory somewhere, cd into that directory, and unpack the
distribution.

If you unpacked, you should be left with the following:

Makefile        - makefile for this distribution
INSTALL         - this INSTALL
README		- a README
SECONDARIES	- a description on how to run a mirror/secondary RIPE database
src/            - source directory containing necessary sources
man/            - directory containing manual pages for executables and
                        configuration file
lib/            - empty directory for database library perl routines
bin/            - empty directory for the executables
doc/            - directory containing some documentation (more coming)
etc/            - empty directory typically used for help and log messages
data/           - empty directory typically used for database files
locks/          - empty directory for lock and pid files
tmp/		- empty directory for temporary files
conf            - sample configuration file

Then, edit the file Makefile to set TOPDIR, LIBDIR, BINDIR, DEFCONFIG and PERL
to the correct FULL pathnames. The best choice is to have the bin and lib
directories right below your top level directory for the database software
(that is, below this directory, automatically included in the archive). If
you set TOPDIR to this directory, LIBDIR and BINDIR will be set
automatically.

LIBDIR  =       directory for database library routines
BINDIR  =       directory for executables
DEFCONFIG=      default configuration file
PERL    =       full pathname of your Perl executable (output from "which perl")

Done this, then you can try:

        make all

This will install the perl library routines in your defined LIBDIR (it
will make the directory if needed), and it will make all executables in
the source directory.  If everything is correct, it should have inserted
LIBDIR, DEFCONFIG and PERL pathnames in the executables.  The default
PERL include path for library routines is altered to take the database
libraries for these executables.  It may give some errors about
directories it cannot create, but these are simply ignored. 

If this succeeded, try:

        make install

It will then put the executables in your defined BINDIR.
It will also make an executable whois client out of whois.pl in the
BINDIR. You'll note that it will again copy the various Perl modules
into the LIBDIR. This does not matter.

Next, have a look at the configuration file for these programs "conf".
There is quite a lot to configure. If you do not want to go through the
bother of changing everything because you just want to test the stuff, make
sure you change EVERYTHING that has path information in it. See the manual
page for dbconfig(5) for all possible configuration commands. Unfortunately
the config reader does not yet check for syntax errors in the config, so
please be a bit careful.

You now have a complete RIPE database setup. If you are familiar with
previous versions of the database you will know your way around, not
too much has changed. You should however read the README file, because
some parts of the indexing have been changed significantly.
